To solve the problem of conventional input-output space partitioning, a new learning algorithm for creating self-organizing fuzzy neural networks (SOFNN) is proposed, which automates structure and parameter identification simultaneously based on input-target samples. First, a self-organizing clustering approach is used to establish the structure of the network and obtain the initial values of its parameters, then a supervised learning method to optimize these parameters. Two specific implementations of the algorithm, including function approximation and forecast modeling of the wastewater treatment system, are developed, comprehensive comparisons are made with other approaches in both of the examples. Simulation studies demonstrate the presented algorithm is superior in terms of compact structure and learning efficiency.

This paper represents a taboo-based neural networks learning algorithm (TBBP) to improve the function approximate ability of neural networks for nonlinear functions. By using taboo search during the search process in the global search area, the algorithm can escape from the local optimal solutions and get a superior global solution form the neural networks. To show the effectiveness of this algorithm, it has been used for approximating different nonlinear functions.

针对目前神经网络在应用中难于确定隐层神经元数的问题，提出了一种神经网络结构的快速修剪算法。该算法在最优脑外科算法（OBS）的基础上，通过直接剔除冗余的隐层神经元实现神经网络结构自组织设计。实验结果表明，快速修剪算法与常规的最优脑外科算法相比，具有更简单的网络结构和更快的学习速度。